Classifications

• • G—PHYSICS
  • G06—COMPUTING; CALCULATING OR COUNTING
  • G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
  • G06T7/00—Image analysis
  • G06T7/0002—Inspection of images, e.g. flaw detection
  • G06T7/0012—Biomedical image inspection
  • G06T7/0014—Biomedical image inspection using an image reference approach
  • G06T7/0016—Biomedical image inspection using an image reference approach involving temporal comparison
• • G—PHYSICS
  • G06—COMPUTING; CALCULATING OR COUNTING
  • G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
  • G06T2207/00—Indexing scheme for image analysis or image enhancement
  • G06T2207/10—Image acquisition modality
  • G06T2207/10016—Video; Image sequence
• • G—PHYSICS
  • G06—COMPUTING; CALCULATING OR COUNTING
  • G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
  • G06T2207/00—Indexing scheme for image analysis or image enhancement
  • G06T2207/10—Image acquisition modality
  • G06T2207/10116—X-ray image
• • G—PHYSICS
  • G06—COMPUTING; CALCULATING OR COUNTING
  • G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
  • G06T2207/00—Indexing scheme for image analysis or image enhancement
  • G06T2207/20—Special algorithmic details
  • G06T2207/20212—Image combination
  • G06T2207/20224—Image subtraction
• • G—PHYSICS
  • G06—COMPUTING; CALCULATING OR COUNTING
  • G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
  • G06T2207/00—Indexing scheme for image analysis or image enhancement
  • G06T2207/30—Subject of image; Context of image processing
  • G06T2207/30004—Biomedical image processing
  • G06T2207/30048—Heart; Cardiac
• • G—PHYSICS
  • G06—COMPUTING; CALCULATING OR COUNTING
  • G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
  • G06T2207/00—Indexing scheme for image analysis or image enhancement
  • G06T2207/30—Subject of image; Context of image processing
  • G06T2207/30004—Biomedical image processing
  • G06T2207/30101—Blood vessel; Artery; Vein; Vascular
  • G06T2207/30104—Vascular flow; Blood flow; Perfusion

Definitions

• the invention reaches the aim with a method for assessing a regurgitant flow through a valve into a moving object from a sequence of consecutive image frames of such object, which images are timely separated by a certain time interval, the method comprising the following steps:
• the invention also relates to a computer product directly loadable into the memory of a computer and comprising software code portions for performing the method as disclosed above when the product is run on a computer.
• Fig. 6 shows a densitometry curve with increasing contrast agent in the left ventricle over time
• Fig. 9 shows two densitometry curves in which the ventricular curve hardly shows any reaction to contrast agent injection: a sign of low regurgitation
• Fig. 12 shows densitometry curves of various regions within the ventricle to be analysed
• Fig. 13 shows an example in which the regurgitation within the left ventricle is visualized by means of a 2D colour map
• Fig. 14 shows an example of the histogram peak shift that can be induced by protocol pixel intensity corrections
• Fig. 15 shows an example of contrast changes due to the presence of contrast agent of previous sessions.
• the invention is particularly advantageous in the quantification of aortic regurgitation based on two-dimensional (2D) angiographic film of X-ray images and it will be mainly disclosed with reference to this field.
• the object of interest is the left ventricle and the valve is the aortic valve. It is however to be understood that the left ventricle, the aorta and the aortic valve can be substituted with any object in fluid communication through a shutting one-way valve.
• the method provides a reliable, reproducible automatically generated quantification of aortic insufficiency based on two-dimensional angiographic X-ray images, which have not been used before for such purpose.
• the new method for Al quantification, based on angiographic images has also the advantage that it blends easily into the current clinical workflow. No extra images beyond what is already acquired during the procedure are required, the patient does not suffer from any additional stress, radiation, or additional procedures and the doctor does not need any additional imaging devices at the operation table.
• a sequence of angiographic X-ray images covering approximately 3 heart beats, shot from a single position is used.
• the images cover the whole or a part of the left ventricle, the aortic valves, any stent inserted, and a small part of the aorta.
• Contrast should be present, and inserted in the aorta at the start of the first heart beat of the analysis. In case of regurgitation, this contrast agent will enter the left ventricle.
• Time-density curves represent a time-evolution of pixel brightness of a sequence of images, particularly perfusion images. For accurate quantification, these curves (also called densitometry curves within the present disclosure) will be created, both for the ventricular area (step 1 ) and the reference area (step 2) as shown in Fig. 12.
• ventricular areas are based on general methods described in literature such as in Gelberg, H.J., Brundage, B.H., "Quantitative Left Ventricular Wall Motion Analysis: A Comparison of Area, Chord and Radial Methods", In: Circulation, vol 59, no 5, May 1979, p. 991 -1000.
• the time density curves within the left ventricle are derived within these sub regions from augmented pre-processed images obtained using the image sequence as input data.
• angiographic X-ray imaging is a projection imaging method
• the left ventricle suffers from background effects for example due to the X-ray absorption of the spinal, or inserted devices such as catheters.
• subtraction will be advantageously performed on the input image sequence before applying the movement correction. This will result in a drastic reduction of background effects.
• the time density computation within the left ventricle sub regions are based on a pre-processed image sequence:
• / is an image / from film /
• l mask is a mask created with the purpose of creating a densitometry image
• Corr is a function which applies the transformation vectors as calculated in step 301 a to an image
• ⁇ i the resulting pre-processed image for the left ventricle which will be used to derive density curves from.
• the protocol correction as described in step 301 b is applied on l y image sequence as calculated above.
• the already present contrast agent has a negative result on the time-density curve(s) of this region. For instance, the regurgitation may seem more severe than the actual case.
• an average wave is calculated for each time-density curve of the frames before contrast injection. This wave is then subtracted periodically from the time-density curve.
• the analysis is based, for example, on the following parameters:
• the period contrast decay in the time density of the aortic area.
• the input for this Al quantification are typically two sequences of pre- process images and the time-density curves derived from those images as described in step 3.
• the measurements are scaled for differences in the amount of added contrast by taking the control contrast area into account.
• heart beats are detected by looking at peaks and valleys in the graphs, as heart contraction will cause a valley, while heart relaxation causes a peak, thus forming a single heart beat together. This detection is further enhanced by taking the heart rate into account in combination with the time between image acquisitions in the image run, or using a linked cardiograph, if any of those are available.
• the penetration of the jet stream is measured by comparing the densitometry values over time and heart beats in different areas in the left ventricle.
• the areas further away from the heart valves (and thus closer to the apex such as region 406 in Fig. 3a) will show decreasing levels of densitometry (see arrows in Fig. 4: curve 401 is lower and relates to an area closer to the apex while curve 402 is higher and relates to an area close to the heart valves).
• densitometry levels close to the apex (region 406 in Fig. 3a) of the left ventricle are high (Fig.
• curve 502 which relates to an area close to the apex, is just as high in densitometry as curve 501 , which relates to an area close to the heart valves region 206 in Fig. 3a), this shifts the quantification towards severe levels of regurgitation, while absence of clear density increases compared to normal background levels shifts the quantification towards milder cases of regurgitation.
• the duration of the presence of contrast agent in the left ventricle is measured by time-analysis of densitometry in the different defined areas in the image sequence (in the example shown in Fig. 12 curve 1202 relates to an area close to heart valves, curve 1203 to an area as middle region, curve 1204 to an area close to the apex, while curve 1201 relates to the reference area). If high amounts of contrast remain during heart beats, the severity of the regurgitation is higher, as the left ventricle has insufficient blood flow away from the ventricle (accumulation of contrast agent, illustrated in Fig. 6, with an increasing trend line 601 fitted to the data).
• the total amount of contrast agent present in the left ventricle, after correction as done in step 302, is directly related to the combined maximal intensities in the left ventricle area(s) and the maximal densitometry in the control area. In case these are more or less similar, this indicates a severe case of regurgitation (Fig. 8, reference graph 802 is very similar to ventricular graph 801 ), as in a healthy situation, no contrast agent should be present in the ventricle at all (minor regurgitation might have a minor effect, as illustrated in Fig. 9, in which ventricular graph 902 hardly shows any reaction to the contrast increase in reference graph 901 ), as all is contained by the heart valves. All cases between these extremes are scaled according to densitometry, again using the comparison of heart beat integrals and integrals over complete curves of ventricular areas compared to the control area.
• Fig. 1 1 illustrates a large inflow compensated for by a larger outflow.
• the parameters considered here include the general increase and decrease of contrast during a heart beat, and the partial integral (the integral taking the densitometry at the start of the heart beat as base level).
• the period contrast decay in contrast fluid in the time-density curves provides an insight in the presence of regurgitation. Only when regurgitation is present, the time-density curve from the aorta shows a period decay of contrast fluid. Furthermore, a period signal within the left ventricular is present in case of aortic regurgitation.
• a Fast Fourier Transform is calculated.
• Per time-density curve the complex magnitude of frequencies corresponding to the cardiac cycle is calculated, by computing the integral using the composite trapezoid method as described in Michael R. King, Nipa A. Mody, "Numerical and Statistical Methods for Bioengineering: applications in MATLAB", chapter 6, ISBN 9780521871587, Cambridge University Press, 201 1 .
• Regurgitation maps can, in fact, be displayed overlaid, for example by varying the opacity, with the input images or with related images, either registered or not, such as those obtained from a different imaging modality, thus providing an immediate overview of insufficiency.
• Regurgitation maps can be presented statically, for example, as the total integral of the time density curve scaled to the reference or dynamically, for example, by displaying the sub integral of the time density curve related to the frame being viewed.
• the representation of time-density curves as colour maps can obviously be achieved independently from any parameter determination and thus numeric indication of insufficiency according to the present invention.

Landscapes

• Engineering & Computer Science (AREA)
• Health & Medical Sciences (AREA)
• General Health & Medical Sciences (AREA)
• Medical Informatics (AREA)
• Nuclear Medicine, Radiotherapy & Molecular Imaging (AREA)
• Radiology & Medical Imaging (AREA)
• Quality & Reliability (AREA)
• Computer Vision & Pattern Recognition (AREA)
• Physics & Mathematics (AREA)
• General Physics & Mathematics (AREA)
• Theoretical Computer Science (AREA)
• Apparatus For Radiation Diagnosis (AREA)

Priority Applications (2)

Applications Claiming Priority (2)

Publications (1)

Family

ID=47358519

Family Applications (1)

Country Status (4)

Cited By (1)

Families Citing this family (14)

Citations (1)

Family Cites Families (7)

• 2012
  • 2012-09-17 EP EP12184678.6A patent/EP2709059B1/en active Active
• 2013
  • 2013-03-26 JP JP2015531491A patent/JP6196309B2/ja active Active
  • 2013-03-26 WO PCT/EP2013/056426 patent/WO2014040758A1/en active Application Filing
  • 2013-03-26 US US14/428,046 patent/US9576360B2/en active Active

Patent Citations (1)

Non-Patent Citations (20)

Also Published As

Similar Documents

Legal Events